Hepatic metastasis by human colorectal carcinoma (CRC) cells requires that CRC must survivethe nitrosative and oxidative stress generated during the first 24 hours after implantationin the hepatic sinusoid. We have established an in vitro co-culture system to study those mechanisms that affect both the hepatic parenchyma and the CRC cell. Our observations reveal that 1) nitric oxide (NO) and reactive oxygen species (ROS) are produced by the host liver as arresting tumor cells create an ischemia/reperfusioninjury, 2) CRC release Carcinoembryonic Antigen (CEA), a 180 kDa glycoprotein, that binds to heterogeneous RNA-binding protein M4 on Kupffer cells (the CEA receptor or GEAR) stimulating release of IL-10 and IL-6, 3) CEA-induced release of IL-10 inhibits iNOS and production of NO and NO- related toxic products, 4) natural mutations in CEA that do not bind the GEAR do not stimulateIL-10 production, and 5) IL-10 promotes CRC survival both in vitro in a liver-CRCco-culture system and in vivo in the liver. However, CRC cells are also predisposed to die as they arrest: 1)CRC cells upregulate intracellular NO and ROS as they enter the circulation, 2) shape change increases gene expression by 2- to 5-fold for apoptotic ligands FASL, RANKL, and death receptors TNFRSF5 (CD40), 7, 8 (CD30), 10B (DR5), and 1 IB, 3) NO increases TRAIL-mediated death of CRC, 4) CEA does not affect intracellularNO or ROS but colocalizes with DR5, and 5) inhibition of CEA expression increases apoptosis but overexpression decreases cell death. Our hypotheses are that 1) CEA inhibits apoptosis in CRC by interfering with death receptor clustering,2) the GEAR facilitates CEA:DR5 interaction and 3) the autocrine effects of CEA within the CRC are more important for metastasis than the paracrine effects through Kupffer cells. These specific aims will test these postulates: 1.) Determine whether CEA requires the GEAR to inhibit anoikis in CRC; 2) Determine whether CEA interacts with death receptors to block initiation of apoptosis in CRC.; 3) Determinewhether the paracrine host response to CEA or the autocrine anti-apoptotic role of CEA in the CRC is more importantfor experimental hepatic metastasis; and 4) Establish whether inhibition of the expression of CEA in vivo by gene therapy inhibits tumorigenicityand metastatic potential in xenograft transplant models. Our primary goal is to develop a gene therapy approach to inhibiting expression of CEA by a Murine Stem Cell Virus-delivered ribozyme or gene silencingconstruct that is eefective in vivo at inducing apoptosis and chemosensitivity.